An NLP Pipeline for Coptic

The Coptic language of Hellenistic era Egypt in the first millennium C.E. is a treasure trove of information for History, Religious Studies, Classics, Linguistics and many other Humanities disciplines. Despite the existence of large amounts of text in the language, comparatively few digital resources have been available, and almost no tools for Natural Language Processing. This paper presents an endto-end, freely available open source tool chain starting with Unicode plain text or XML transcriptions of Coptic manuscript data, which adds fully automatic word and morpheme segmentation, normalization, language of origin recognition, part of speech tagging, lemmatization, and dependency parsing at the click of a button. We evaluate each component of the pipeline, which is accessible as a Web interface and machine readable API online.